Skip to content

AI LLM API

One endpoint, one API key, more than 20 models from Anthropic, OpenAI, xAI, Google, DeepSeek and Moonshot. You top up in VND — no international card and no separate account at each vendor.
Base URL https://revidapi.com/v1
Auth Authorization: Bearer sk_... or X-API-Key: sk_...
Formats OpenAI, Anthropic Messages, OpenAI Responses
Billing unit credit — 1 credit = 55,8 VND

Three protocols, one key

Pick whichever matches the library you already use:

Path For Note
POST /v1/chat/completions OpenAI SDK, LangChain, n8n, Dify Most common
POST /v1/messages Anthropic SDK Keeps system and block content
POST /v1/responses Codex CLI, Agents SDK Proxied as-is, tool calls intact

GET /v1/models returns the live catalogue with prices.

First call

curl https://revidapi.com/v1/chat/completions \
  -H "Authorization: Bearer sk_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-5",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

With the OpenAI SDK, only base_url changes:

from openai import OpenAI

client = OpenAI(api_key="sk_YOUR_KEY",
                base_url="https://revidapi.com/v1")

r = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[{"role": "user", "content": "Hello"}],
)
print(r.choices[0].message.content)

Free trial

New accounts get 10 free calls on the Gemini trial models. No credits deducted, no top-up needed.

Billing

Billed per token, input and output priced separately. See Models and pricing.

  • Every call is charged a minimum of 10 credits.
  • Failed calls are refunded automatically — you never pay for an answer you did not receive.
  • Cached input is far cheaper than fresh input. See Saving with cache.
curl https://revidapi.com/v1/credits -H "X-API-Key: sk_YOUR_KEY"

Next